 OK, now green, yeah. So it's a very broad topic. So I decided after seeing Simona's topic on monsoon and tropical connection, so I thought maybe I should narrow this down to tropical ocean atmosphere feedback. So we'll do that. And then the second half of my lecture is going to continue somewhat where Simona has left. Namely, we want to look into the internal variability of the summer monsoon of Asia mostly. So maybe a little bit strange, but maybe a healthy change in the approach is I'm not going to show any equations. I'm not even a model experiment. So I'm going to look mostly into observations because I feel with accumulation of much research, you can actually see a lot of physical processes and mechanisms operating in real world ocean and atmosphere. So that's what I'm going to do. So I know I understand this is a rather theoretical oriented summer school. But maybe sometimes it's good to have some reality check. So yeah. So although Simona and I are both talking about monsoon, you will see the approaches and even maybe some of the views might be quite different. But that reality is always much more complicated and much more beautiful than we can ever imagine. This is what Alesson I learned from doing this line research for 23 years. So anyway, OK. So this is actually historical compilation of satellite observations down by NOAA in the early 70s. So I guess the first artificial satellite was launched by Soviet Union in 1957 or eight. And then the US put a few satellites in orbit and they mounted cameras on those satellites and taking pictures of the Earth. So this is like a four-year compilation of satellite surface brightness, the Earth's brightness composite for January of those four years. So I think there are many features one can enjoy attention to. But maybe the two particular features I think are very striking and especially in terms of puzzling from the spatial distribution of solar radiation. For example, it's in January. So over the continent, southern hemisphere, you see a lot of deep convection making the clouds very bright. Whereas over much of the stretch of the equatorial Pacific and Atlantic, ITCZ in winter, northern winter, still displays into the northern hemisphere. And also, so that's kind of strange. Just given the solar radiation averaged over the annual cycle, it should be mostly symmetrical by equator. But then the climate is clearly asymmetrical. And another thing is in the western Pacific and the maritime continent, you see a lot of deep convection. Whereas in the eastern half of the equatorial Pacific, it's all clear. So no convection whatsoever. So that's also unexplainable from the solar radiation because solar radiation is only uniform. So it must be due to something else, and in particular due to the interaction between the ocean and atmosphere. So this is a similar depiction of rainfall distribution in gray shade. And then now the SST, I choose high SST contours above 27 degrees Celsius to highlight the co-location between high SSTs and high rainfall. So I guess without SST variations, the convective activity can be very confusing. Like the Madden-Johlin oscillation, people show that convection is organized into a big blob and a big blob tend to propagate eastward. So the dynamics of that is a bit complicated. But in the presence of a strong SST variations, the meteorological problem becomes easier because you simply just say for whatever reasons, it's just a high SST tends to anchor deep convection. I mean, this is understandable. Warm SST allows a big, moist static energy in the boundary layer. They are more convective, unstable, and so on. So if you are a meteorologist, so you would conclude, oh, I understand, because why the ITC stays north of Equator is because SST is warmer in the northern hemisphere. It's straightforward. But if you are oceanographer, you will say, oh, SST is warmer north of Equator because your rain-band conversion song is north of Equator. So clearly, there's a chicken and egg circular argument between the meteorologists and oceanographers. So when we hear such circular argument that's indicative that you are facing a coupled problem between the ocean and atmosphere, so you have to, you know, solving the chicken problem or egg problem isn't just going to be enough. You have to solve a coupled chicken egg problem. So that's the example to illustrate the coupled approach to climate research. I guess this became a popular since about 1980s, so the beginning of so-called a couple of ocean atmosphere dynamics research era. And then coming back to this North-South asymmetry problem, so clearly, you would like to have some positive feedback that would generate North-South asymmetry. So one possibility is, perhaps, is considering in the tropics, easterly winds are prevailing. So suppose, for some reason, you can get the northern hemisphere slightly warmer than the southern hemisphere. And then you are going to drive winds from a colder hemisphere into warmer hemisphere like here. And then the southerly winds are going to turn due to the Coriolis effect towards the east. So you gain a westerly zonal wind component due to the cross-equitile wind. And then similarly, in the southern hemisphere, the Coriolis parameter is negative. So you turn the winds to be easterly to gain an easterly component. And then superimposed on the background easterly winds, you're going to conclude over the warmer water, the easterly winds are weakening, whereas over the cold water, the easterly winds are strengthening. So if you invoke the wind dependency of the surface evaporation, so high winds will evaporate more water from the ocean surface, therefore, to cool the ocean surface, like vice versa. So that would, if you invoke evaporative effect, then this reduced wind speed is going to intensify the initial warming, whereas the enhanced easterly winds are going to cool the southern hemisphere even more. So as you can see now, there is intrinsic positive feedback between the ocean atmosphere through evaporation that tends to destabilize the symmetrical state of the Earth's system. So these nowadays we call this wind evaporation SST feedback, the West feedback. As you will see, it's actually one of several important feedback mechanisms between ocean atmosphere, driving common variability and distribution in the horizontal direction. So just as a reality check, if you go to Eastern Pacific or Atlantic, you will notice the southern hemisphere, the easterly winds are much higher than in the northern hemisphere, where the tree winds converge. So in the southern hemisphere, like 10 south, the typical wind speed is about 8 meter per second, whereas in the ITCC, the winds are highly variable. You still have like thunderstorms and so on. So the vector winds might be vanishing, but the scalar wind, so it's a scalar wind that is important for evaporation will remain to be some small value, say 5 or 6 meter per second. So the observation clearly shows there is a north-south asymmetry in the wind distribution, and therefore the wind speed, and therefore in support of this West feedback. And then this West feedback, of course, now I'm seeing conveniently introducing warmer SSC anomalies in the northern hemisphere. I could do the same for cooler SST. I will still have West feedback operating to amplify my cooler northern hemisphere. So it doesn't help. It just say the feedback that would amplify the asymmetrical perturbation. It doesn't say why northern hemisphere, but never southern hemisphere. So I guess a popular theory, prevailing theory now, is the AMARC. So AMARC would drive transport heat into northern hemisphere. And that would require the atmosphere to transport the energy back into southern hemisphere, so that Simona and others have talked about. So that would require to displace the ITCC into northern hemisphere for the atmosphere to transport energy back into the southern hemisphere. So this is a global view on the zonal main asymmetry of the system. But obviously, you can point to the Indian Ocean, for example. Indian Ocean has a strong seasonal cycle. You don't even see whether it's in the northern or southern hemisphere. But for some reasons, the Pacific and the Atlantic seem to favor following this argument. So in addition to this global energy framework that dictates the zonal main circulation of the atmosphere, there are also other ideas about why Pacific, why not Indian Ocean. So the first time I remember was when Kirk Bryant told me, oh, you just tilted the coast of South America. So if you do that, then the prevailing traits are going to cause upwelling in the southern hemisphere because the acumen flow goes southward, leaving the coast. Whereas the downwelling would occur north of the equator. So that asymmetry in upwelling would favor a warmer northern hemisphere than the southern hemisphere. So this is a more like a local tropical trigger of asymmetry to a particular basin. So people have done experiments like this. Everything is symmetrical in a couple model. And then you just simply tilt the coast of the eastern continent. You're going to cause a strong asymmetry to the west across the entire basin. So the SST contours shielding rainfall and winds in vectors. So you can indeed, by modifying tropical continental geometry, you can end up with asymmetrical state in the basin to the west of the continent you manipulate. So all those things, all make sense. So which is more dominant in what sense? So all those things, I think, still need to be worked out. So is this also the case? I mean, the Pacific wasn't really the case. I guess also the Atlantic, that's the case? Yeah, the West Africa stick out. So that also you can do experiments. Indeed, you can cause asymmetry. OK, so this is kind of an introduction to this west feedback and also the global control versus the more eastern boundary effect, local effect, if you will. And then I would like to change to the next topic. So I said from the silo image, the western Pacific is highly convective, whereas the eastern Pacific is free of deep convection. So why is that? So that requires invoking this so-called equatorial upwelling concept. So I said in the tropics, the easterly winds are prevailing. So in the presence of the easterly wind, slightly away from equator, you have a Coriolis force and the acumen flow goes forward, both north and south equator, and then will drive upwelling. Because the surface of current divergence will call for upwelling over the cold water from beneath. As John and others have shown, abyssal water is near freezing, so enough cold water. You just need to pump up. John mentioned people in the hot weather, they want to pump some cold water from beneath. Nature is doing this for us in the eastern Pacific, at least. So this is an SST distribution across Pacific in Enumé. So in the eastern Pacific, the SST is something like a 24-ish in the eastern Aquatouille Pacific. So here's the island chain, Galabagos Islands, right? So how many of you know there? So I know like Galabagos, there live a lot of exotic animals. So there's maybe one animal might be surprised to you. So namely, on the equator, there are penguins, right? Galabagos penguin. So they live on equator, so that's just remarkable. It's just that the upwelling is strong enough to make those penguins comfortable, at least, right? That's quite remarkable. I just thought it's just the natural pump system that keeps the Galabagos in the comfort zone because penguin people are associated with Antarctic. It's extremely cold. It's not as cold as Antarctic, obviously. Yeah, so anyway, so the east three winds, so in a sense, there's positive feedback. As long as you can make the eastern ocean cooler, then the temperature gradients are set up to drive strength in the east three winds. So there's a positive feedback. So this is actually what Jacob Bjorkner's envisioned for El Nino Southern Oscillation. So we'll come back to that a little bit. So this is, I assembled those three plots together to get kind of a more comprehensive view of the tropical climate, both at the surface and also in the upper part of the ocean. So this is something similar to what I shown earlier, SST together with rainfall. And this is a Z20. So 20 degrees, also some depths in the ocean. So deep in the west and shallow in the east and the surface of wind velocity distribution. And this is the upper ocean stratification. So another effect of blowing the eastern wind is it set up, it calls for, it set up the pressure gradient to balance out the eastern wind stress. So that calls for thermocline to tilt this way. So in other words, the surface currents are driven this direction. Therefore, they accumulate in the western Pacific, whereas the eastern Pacific is losing water. So the thermocline has to show. So that's also very important. So the total of the thermocline in the Pacific determines why the cooling occurs in the eastern ocean but not in the western ocean. Because the upwelling occurs across the Pacific ocean. But only in the eastern ocean, when the thermocline is very close to the surface, you can very effectively pump the cold water from beneath. Well, here, if you pump, set up a pump at 50 meters, you still pump a very warm temperature from beneath. So it's not going to help you to cool. But so in that regard, both the upwelling and also the depth of the thermocline are both important in deciding the surface temperature you're interested in. So this is a biocannous feedback. So the upwelling will tilt the thermocline to the east and then that will cool the eastern Pacific. And the cooling of the eastern Pacific is going to push the convection westward, intensifying the eastern wind. So this is a positive feedback between the zonal wind and the sea surface temperature through the upwelling and the total of the thermocline. So at this point, we have seen two major positive feedbacks in the tropical ocean. One is the biocannous feedback. This is a very strong feedback, as we will see in oscillation between El Nino and La Nino. Another is the west feedback. Their favor tends to amplify the asymmetrical perturbation in the meridional direction. We will see this west feedback in another real example just a bit later. So OK, so this is the annual mean. So in Simone's talk, clearly, the north-south, the seasonal cycle in solar radiation, I guess we just passed summer solstice, longest day in the northern hemisphere. So obviously, summer is warmer because more solar radiation, longer day, and so on. So you can probably imagine like the north-south asymmetry is going to be much stronger in northern hemisphere summer to reinforce the annual mean asymmetry. Whereas in the northern hemisphere winter, the solar radiation is more intense in the southern hemisphere than in the northern hemisphere. So naturally, the north-south asymmetry might be reduced or maybe even reversed as in many climate models. So we will see what happens to that. So this is, again, an observation for the eastern tropical Pacific in the 120 degree west. So that's something like 4,000 kilometers away from the coast of South America. So again, here at SST in contours, and then rainfall precipitation in this blue shade, and then the white contours are the bigger values of rainfall and the surface winds in vectors. So the surface winds obtained from ship observations, so is SST, rainfall is based upon satellite estimates. So indeed, as we discussed earlier, in September, for example, the northern hemisphere ocean at least is at the height of the summer, whereas the southern hemisphere is in winter. So as you might imagine, the southern hemisphere is a lot cooler than the northern hemisphere. So a very striking ITC displays to the northern hemisphere in September indeed. So the ITC is here, Galapagos Islands here, Hawaii here, Peru here. So Peru is very dry because both coastal upwelling and equatorial upwelling clears up all the convection in the southern hemisphere. So this is this. So September, northern hemisphere summer asymmetry strongly developed. So very strangely, perhaps, the nature is kind of really elegant in the sense in March that the southern hemisphere is in high summer, at least as far as the ocean is concerned. So the southern hemisphere SST temperature would rise, whereas northern hemisphere temperature would drop. So that's the time actually a nearly symmetrical climate emerges in the eastern equatorial Pacific Ocean. So this is what you see like the rain band, the kind of symmetrical two rain bands each in one hemisphere. So this is kind of an elegant delicate in the sense the nature reduces the north-south asymmetry but not to eliminate. So it just reduces the asymmetry to a degree that becomes symmetrical, but nature doesn't push the asymmetry to push the southern hemisphere to be too warm to support the southern ITCZ while eliminating the northern ITCZ. But in models, it's different. In models, in March, the northern ITCZ disappear in the southern hemisphere. There is a very stubborn southern ITCZ in March in models. So that has been a problem, first identified in about 1990s, so 20, 25 years. Afterwards, not much progress has been made. So models still suffer this. People call it like a double ITCZ problem. So yes, for some reason, the nature just stops at the right magnitude to have a beautiful symmetrical ITCZ emerging in the eastern equatorial Pacific. So this is a very in nature double ITCZ appears only briefly during about two months between February to April. So every year, if you go to see satellite pictures in March, April, and maybe late part of February, you will see this double ITCZ. It's just tremendously. When first people spotted, I think 1967 or something. So when people first, the camera sitting on the satellite took many pictures over like March, 1967. And then they did a composite. Beautiful double ITCZ appear. So various people have a different, various theories, including Chani may propose the one. But it turned out not to be too relevant. It turns out SST is the real anchor, not really fancy dynamics of the atmosphere. So that's another example. SST can provide a shortcut for deep connection to organize just because SST determines the surface more aesthetic energy. So it's so powerful distribution that the other things have just to yield to SST effect. OK, so this is maybe a brief summary of what I just said. So the Eastern Pacific and Atlantic Ocean develops very strong North South asymmetry. So we now believe continental geometry, both global and tropical local. So I guess John and Brian and Dave's presentations have shown how the global geometry, especially the opening of the southern ocean, appears to be very powerful to shift the AMOG to have a sinking branch in the northern hemisphere. And also the tilt of the South American continent, the tropical continental geometry might also have an effect in determining which basin would have asymmetry and which basin might not have. So that's that. And also the east-west asymmetry is for the east-west asymmetry, the Biaquines feedback is strongly involved in creating the coal town and therefore allowing Globogus penguins to thrive in those even on the equator. OK, so now I'm going to switch gear a little bit about variability. So I guess in standard discussion of tropical-LSE interaction, you would start from El Nino southern oscillation. And that's indeed what history took place just like in this sequence. So people discovered the couple ocean-atmospheric interaction first by studying El Nino southern oscillation. So I just thought this is kind of a very striking example. Like seasonal variations, of course, the sun is driving the variation, SST and winds and circulation and so on. But across the Pacific Ocean, if you go to measure SST on the Globogus islands, you will find the SST would fluctuate from something like minus 1 degree to plus 1 degree or even more fluctuating on intangible time scales. And then if one goes to Darwin, Australia to measure atmospheric pressure, you will find that pressure also fluctuates on both weather timescale as well as intangible timescale. And what's really remarkable, if you have two time series to plot together, you immediately see there's a high correlation between the two. One in the western Pacific atmospheric pressure. One in the eastern Pacific ocean temperature. Why the two should even have a relationship, right? So I thought this was really bizarre and remarkable. So anyway, through the research in the 80s and 90s led to understanding now that we would approach this problem from the Bjarkeans feedback perspective. So here I'm showing composite sea surface temperature anomaly during September to November of an annual event and in color and surface rainfall anomalies in black solid contour positive and dashed black contours or negative contours. And then here I'm showing like some of the depth here measured by 15 degrees after some of the depth anomalies in color here and surface wind anomalies in vectors here. So the thermoclamp, the flattening of the, so let's start from the eastern Pacific warming. For some reason, you can warm the eastern Pacific. Now you will drive the surface of westerly wind anomalies into the eastern Pacific. And that will do two things. One is to reduce the upwelling in this region. Another is to flatten the thermocline. So remember the thermocline is tilted originally like this. And if you relax the winds, the thermocline has to flatten. So the flattened thermocline would deepen the thermocline in the eastern Akatari Pacific. So even if you have the same pump, the water you are pumping up is going to be a lot warmer than in normal conditions. So that illustrates the thermocline depth adjustment in the ocean is a critical element of El Nino southern oscillation. So if you really look carefully, so like in the eastern Akatari Pacific, the winds at this stage actually doesn't change that much. So the pump is not changing that much. But it's the water the pumps takes up. It's warming up much more than in normal conditions. So this clearly has a positive feedback through the tilted thermocline. It would amplify the surface warming in the eastern Pacific initially start. So this is a kind of a just a traditional Biakenis view of El Nino southern oscillation. So this is, I mean, I thought remarkable in a sense. The physical understanding of all those processes led to a successful seasonal forecast of El Nino southern oscillation. And we don't have time here to talk about El Nino effects in remote regions. Maybe in the second part of my talk, I will reinforce this El Nino influence on Asian monsoon, summer monsoon. But El Nino basically has a global influence. So it's so whatever you study, wherever you live, you will see El Nino signals. So in a sense, if you're a researcher, it's a convenient clock. You can time against your data to explain why your data has this variation. OK, so this is El Nino. I think all this you might have learned in your lectures, classes, and so on. So the last bit, however, I think is quite surprising to myself. And I believe it's something you probably have not seen before. So I will try to walk you through this quite for myself. I've been doing this for like 20, 25 years. It's still sharp. I was very surprised by the results. So anyway, you all started from this rainstorm event in Peru, March 2017. So if you follow the news, that's a big deal. So in Peru, in this part of Peru, annual rainfall is something like 75 millimeter. Almost never rain, because of uproiling and so on. But in a single, like three months of 2017, they observed 631 millimeter. So that's eight times the annual cometology. So you can see people devastated. So this is not terribly surprising for people in Peru. In a sense, when El Nino happened, especially big El Nino happens, they tend to have big storms like this. But that's why El Nino is a term actually used by Peruvian fishermen to indicate the coastal warming. And then Biaquines and subsequent research show that this coastal warming is actually basin-wide phenomenon involving a large-scale ALC interaction. But what's bizarre about this El Nino event is the Nino 4 region indicates the central Pacific equatorial. Nino 3 indicates the eastern half. And Nino 1 plus 2 is this Ecuador-Peruvian coastal region. So if you look at the coastal surface temperature, the anomalies reaches something like 3 degrees Celsius. So at the warmest point, the temperature exceeded 29 degrees Celsius off the coast of Peru. It's just tremendous. So this is kind of the cause, in other words, the coastal warming. But then, if you look at like a coastal El Nino, it's not a coastal, so it's a basin-wide phenomenon. But if you go to central Pacific, eastern Pacific, you'll find actually the SST anomalies, somewhat negative, a weak El Nino state. So this is really bizarre, in a sense, why on the basin scale, you have a weak El Nino. But yet, on coastal Peru, you've got 29 degrees Celsius water with a downpour 10 times of the comatology called rainfall. So this is just very strange. So we decided to take a look. So this is a total rainfall observed during March 2017. So you kind of see there is a double ITCC with two rain bands and so on. But the southern bands, southern ITCC, clearly it's more intensified than northern ITCC. So if you do like a subtract long-term comatology and so on, you see like a positive rainfall anomaly of coastal Peru and a negative one. But what's really strange is, I said, on the basin scale, it's La Nina. But really, this dipolar rainfall pattern is actually somewhat basin scale. It's not limited to the coast, clearly. So this is a really bizarre phenomenon. Then if you go to see SST, so SST like this, total SST. And then this is the extreme warming on the coast. But SST, even SST, you see like an eastern half the Pacific, the southern hemisphere is warming, then the northern hemisphere. So still indicating, it's not like just a convection doing random things. The convection is following the order of SST. So this is what happened in 2017. So as I said, March is a special season that the eastern Pacific becomes nearly symmetrical about equator. So El Nino itself peaks like in northern winter, November, December. So people tend to study El Nino in the winter season more, in the peak season. But then March, I thought March is a different season. So maybe we should look at March, April season more closely separately. So then we did an EUF analysis in this domain of rainfall, rainfall variability. And then this is EUF1. EUF1, you kind of see the kind of blue, greenish color is increased rainfall, and the blinds decreased rainfall. And then the surface winds and SST anomaly. So SST anomaly is positive across the basin for the first mode. And if you look at the time series, you immediately notice it's dominated by 83 and 98. So extremely El Nino. Extremely El Nino can push the warm water all the way across the Pacific Ocean, triggering convection, even on the equator. So this is extremely El Nino. And then the second mode is related to the Peruvian flooding. So you see the second mode is dipolar, meridional dipolar. The northern hemisphere rainfall generally increased, whereas the southern hemisphere rainfall decreased. And then you see the winds are blowing from a dry hemisphere into warmer hemisphere. And then the winds also turn in this sea shape. So that's indicative of West feedback. So in other words, in the Pacific, I never expected to see West mode in terrestrial variability in the Pacific. I just thought El Nino is so dominant. So you won't give West a chance. But you still see this in limited seasonal window for February, March, April. And what's really surprising is that UF2 is highly correlated with central Pacific SST anomaly. So this is really bizarre. Somehow this is a, some El Nino would have a meridional dipolar pattern in March, April. And that could cause heavy rainfall in the southern hemisphere. So this is bizarre, because El Nino is generally considered to be a pattern peaking on the equator, symmetrical about the equator. Whereas now I'm saying like in March, April, it's a dipolar meridional asymmetrical. So there's something we don't quite understand. But then if you go to really look at how this meridional dipole develops, so we know how the dipole looks like in March, April. We just go back to one season to see what happened. To the SST in November, December, January season. So this is the SST anomaly in color shading and surface winds in fact. And then this is the rainfall anomalies here. So here in the solid contour, I'm showing SST, chronological SST, above 26 degrees Celsius. So you can see like in November, December, January, the main climate is highly asymmetrical. Warm water is confined in the northern hemisphere. So even if the SST warming is symmetrical, the rainfall response is going to be asymmetrical. So only northern hemisphere is warm enough to have a convective response. Southern hemisphere just doesn't say what the ocean is doing at this point, because the main climate is so asymmetrical. So that determines the winds actually are highly asymmetrical, blowing from southern hemisphere into northern hemisphere. So SST is nearly symmetrical, but atmospheric anomalies are already asymmetrical about the equator. So you can imagine in the next season, when my main climate becomes asymmetrical, this asymmetrical atmosphere response is already planting seeds of asymmetry in the system. So in the next season, you will see, oh, OK. So this is a meridional overturning circulation. So this is a transect across eastern equatorial Pacific, 15 south, 15 north. So in the northern hemisphere, El Nino would cause upward motion because of strong convection. Whereas in the southern hemisphere, no counterpart in the deep circulation and adjustment. So in the next season, when we go to February, March, April, when the climate is nearly symmetrical, so the ocean is warm enough and conducive to convection in both northern and southern hemisphere. So that's why northern hemisphere remains a positive rainfall anomalies. But now you develop a negative anomalies to the south. And the winds are accelerating. So the tree winds, the south-easterly tree winds are accelerating in the southern hemisphere and decelerating in the northern hemisphere. So this is kind of a classic West feedback type arrangement. So it's not like the West feedback is not working. But you need a right environmental condition for this to take place. It can transform a symmetrical El Nino into anti-symmetrical meridional dipolar structure. And now if you go to see the vertical motion in the atmosphere in the eastern Pacific, you see both in the northern hemisphere and southern hemisphere, you see very deep circulation anomalies consistent with the convective anomalies. So here a humidity increase, and here humidity decreases. Yeah, so this is quite surprising to me, although I've been around for quite a while. But still the nature would often surprise us with something new. And after you think through it, it all makes sense. So if in the, so that means the SSTs and the Nino 1 and the Nino 4 region, they are anti-correlated in March? So that part is more complicated. So although we motivate, our study is motivated by Peruvian flood, right? But when we go back to correlate the Nino 4 with the Nino 1 plus 2, the coastal SST, statistically they are not correlated. So another surprise in way. So I haven't figured out that part yet. But I just thought it's quite bizarre. El Nino would develop into an anti-symmetrical pattern when the conditions are right. Yeah, so this is just to illustrate this meridional oscillation in the Hoffman-Moller diagram. So here from 1979 to current and equator northern and southern hemisphere, somewhat the zonal average in the eastern equatorial Pacific. So you see like a, cum logically, you have both ITCZ, both rain bands in each hemisphere. And then occasionally, the big El Nino would intrude into eastern equatorial Pacific, causing rainfall, right? Especially 83 and 98. And then you see that like in each hemisphere, the rainfall would oscillate. And rainfall oscillation actually occurred in the north-south anti-symmetrical fashion. So this is a scatter plot between the southern rain band and northern rain band rainfall. So you see this rainfall intensity between northern and southern ITCZ is anti-correlated. So sometimes you only got southern ITCZ such as in 2009. And in the immediate next year, you only got northern ITCZ, not southern one. So now I think we kind of figure out this is the meridional oscillation is actually controlled by El Nino southern oscillation. Very bizarre. OK, so just put up a summary of the second half. So the second half is more like the current research. So we find like we call the eastern Pacific ITCZ dipole. It's a west mode confined to March-April season when the main state is symmetrical. And therefore, the atmosphere feedback is strongest. So you can choose the northern or southern hemisphere, or both to have a convection. And it represents internal variability in relative intensity between northern and southern ITCZ. And strangely enough, this epic mode is preceded by moderate El Nino southern oscillation phenomenon as indicated by Nino 4, central Pacific SST anomalies. So this paper just published in Journal of Climate. If you're interested, you can take a look. So I'm just still five minutes, but I like to open for questions for both the first part, kind of a general introduction to tropical climatology and major positive feedback between ocean atmosphere. And the epic mode just show both are present in the different stages of El Nino southern oscillation. OK, questions? Yeah, if I.